Single Cell Oil production: a new approach in biorefinery

Thesis by

Gaetano Zuccaro

Ph.D.

in

INDUSTRIAL PRODUCT AND

PROCESS ENGINEERING

(XXIX)

University of Naples Federico II

School of Polytechnic and Basic Sciences

Department of Chemical, Materials and Production

Engineering (DICMaPI)

1

Single Cell Oil production: a new approach in

biorefinery

by

Gaetano Zuccaro

Tutor: Prof. Domenico Pirozzi

Co-Tutor: Prof. Giuseppe Toscano

Prof. Antonino Pollio

Prof. Gabriele Pinto

Dr. Jean-Philippe Steyer

Dr. Robert van Lis

Course coordinator: Prof. Giuseppe Mensitieri

2

Penso che sia necessario educare le nuove generazioni al valore della sconfitta.

Alla sua gestione. All’umanità che ne scaturisce.

A costruire un’identità capace di avvertire una comunanza di destino, dove si può fallire e

ricominciare senza che il valore e la dignità ne siano intaccati.

A non divenire uno sgomitatore sociale, a non passare sul corpo degli altri per arrivare

primo. In questo mondo di vincitori volgari e disonesti, di prevaricatori falsi ed opportunisti,

della gente che conta, che occupa il potere, che scippa il presente, figuriamoci il futuro, a

tutti i nevrotici del successo, dell’apparire, del diventare.

A questa antropologia del vincente preferisco di gran lunga chi perde.

È un esercizio che mi riesce bene.

E mi riconcilia con il mio sacro poco.

(P. P. Pasolini)

3

ABSTRACT

A socially responsible economic growth, devoted to the future generations, requires long-term

secure and available resources for industrial production, in terms of raw materials, energy and

water. It should be environmentally friendly and a forward-looking financial system capable

of future challenges with a global point of view. However, the present economic model based

on the non-renewable fossil resources (oil, natural gas, coal, minerals) for energy and

industrial production is the reason of energy instability, climate changes and therefore it

cannot be considered sustainable.

In this context, biotechnological techniques, such as the biorefinery are becoming more

attractive. The biorefinery consist in the sustainable transformation of biomass, such as plant,

algae, yeast, bacteria, into a wide range of marketable products and, in the mean time, into

energy. Oils derived from biomass sources are named Microbial Oils, Unicellular Oils or

Single Cell Oils (SCOs). The sustainable production of Single Cell Oils (SCOs) has garnered

recent attention. The goal is new strategies or the biochemical/microbial conversion processes

in order to increase their productivity and competitiveness. The identified strategies include

the metabolic and genetic engineering of microorganisms, new fermentation technologies,

innovative process choices, low-value feedstock and the recycle of by-products. SCOs are

produced by oleaginous microorganisms which are able to accumulate between 20% and up to

80% lipid per dry biomass in the stationary growth phase under nutrient limitations, e.g.,

nitrogen or phosphorus, with simultaneous excess of carbon source. Depending on the

oleaginous microorganisms including bacteria, yeasts, microalgae or fungal species, fatty acid

profile of SCOs can vary making them highly suitable for different industrial applications.

The SCOs, obtained from plant and microbial sources, offer several advantages, including

faster production, less labor, more season and climate flexibility, and easier scale-up.

The aim of this thesis is to explore the capacity, efficiency and productivity of oleaginous

microorganisms, grown on agro-industrial lignocellulosic biomasses, to accumulate Single

Cell Oils. This allow to reduce the environmental problem associated with the recycle of

residues of various industrial processes allowing while increase the economic advantage

linked to the SCO production. In order to make Single Cell Oil production more economical

and sustainable, the experimental activities were aimed to achieve the following objectives:

The optimization of operating conditions of enzymatic hydrolysis evaluating the

effects of enzyme concentration, temperature and pH on the fermentable sugar

production. The experimental tests were carried out to assess the viability of

4

innovative fermentation processes, i.e. in single stage or single reactor, with the aim to

reduce the capital costs. It was found the synergism of enzymatic mix as well as the

positive effects related to a lower temperature, suspended composition of

hydrolysates, pH control and reaction time.

The implementation of enzymatic hydrolysis and oleaginous fermentation in single

reactor (SRF) that offers an useful option to integrate in a single reactor two different

stages of the microbial oil production: the enzymatic hydrolysis of the pretreated

lignocellulosic biomass and the microbial fermentation of the obtained fermentable

sugar mixture. Specific glucose consumption rate (μs), lipid yield (Ylipid) were

calculated. The results suggest positive potential application of such process that still

remains unexplored for Single Cell Oil production but demonstrate that they are

suitable for biodiesel, bioplastic production or for other products of industrial interest.

The investigation of synergistic effect of yeast-microalga mixed cultures, in order to

find suitable operating conditions to improve SCO production in mixotrophic

microbial cultures. To test the consortium in clearly defined conditions, a synthetic

medium was developed that integrates necessary elements of known culture media for

both organisms, with the use of pH control. Growth series were done in batch, under

constant light, agitation and temperature, and monitored gas exchange. In the model

system, symbiotic growth was observed of the consortium with synergistic effects on

biomass yield. Similar results were obtained in a system using lignocellulosic

hydrolysate, except for an increase in lag phase due the presence of inhibitors. Growth

even in anaerobic conditions (N2) confirmed synergistic interactions between the

microalga and the oleaginous yeast. The performance of the consortium under

different conditions is discussed in terms of growth rate, biomass production and lipid

content of the biomass.

The implementation of yeast-microalga mixed cultures in open pond, overcoming

contamination problems. The yeast grew, reaching the maximum concentration just

after few days. The rapid growth kinetic was expected for the yeast while the very low

cell proliferation can be explained by the low quantity of organic carbon in the

medium.

Keywords: Lignocellulosic biomass, Single Cell Oils (SCOs), steam explosion, enzymatic

hydrolysis, oleaginous yeasts, microalgae, mixed cultures, FAME.

5

PREFACE

This thesis describes the most important results achieved during my Ph.D. study. The work

was carried out mostly at Biochemical Engineering laboratory, Department of Chemical,

Materials and Production Engineering (DICMaPI), University of Naples Federico II in the

period from February 2014 to February 2017. Part of experimental activity was also

conducted at the Bioscience and Territory Department, University of Molise in Isernia, Italy;

at the Department of Biological Science at the University of Naples Federico II, Italy; and at

INRA-LBE (Laboratoire de Biotechnologie de l'Environnement) in Narbonne, France. The

thesis was submitted to the Department of Chemical, Materials and Production Engineering

(DICMaPI), University of Naples Federico II on April 10, 2017 as part of the requirements of

the academic degree of Doctor of Philosophy (Ph.D.) in Industrial Product and Process

Engineering.

Gaetano Zuccaro

Napoli, April 10, 2017

6

CONTENT

ABSTRACT 3

PREFACE 5

LIST OF ABBREVIATIONS 8

1. INTRODUCTION 11

1.1 A GLOBAL REDESIGN? AMBITION OF THE “CIRCULAR ECONOMY” 11

1.2 THE DEFINITION FOR BIOREFINERY 11

1.3 SINGLE CELL OILS 12

1.4 PROPERTIES OF LIGNOCELLULOSIC MATERIALS 13

1.5 PRETREATMENT TECHNOLOGIES 16

1.6 ENZYMATIC HYDROLYSIS 17

1.7 OLEAGINOUS MICROORGANISMS 20

1.7.1 OLEAGINOUS YEASTS 20

1.7.2 MICROALGAE 23

1.8 CULTIVATION STRATEGIES 26

AIM OF THE THESIS 28

2. MATERIALS AND METHODS 29

2.1 PRETREATMENT OF LIGNOCELLULOSIC MATERIAL 29

2.2 THE HYDROLYSIS OF LIGNOCELLULOSIC BIOMASS 29

2.2.1 ARUNDO DONAX HYDROLYSATE 29

2.2.2 WHEAT STRAW HYDROLYSATE 31

2.3 SEPARATE HYDROLYSIS AND FERMENTATION (SHF) AND ENZYMATIC HYDROLYSIS AND

OLEAGINOUS FERMENTATION IN SINGLE REACTOR (SRF) OF STEAM EXPLODED ARUNDO DONAX

31

2.4 MIXED OLEAGINOUS YEAST-MICROALGA CULTURES 33

2.5 COMBINED YEAST AND MICROALGAL CONSORTIUM IN PILOT SCALE RACEWAY POND FOR

URBAN WASTEWATER TREATMENT AND POTENTIAL SCO PRODUCTION 35

2.6 ANALYTICAL METHODS 37

3. RESULTS AND DISCUSSION 40

3.1 PROCESS PARAMETER OPTIMIZATION OF ENZYMATIC HYDROLYSIS 40

3.2 SEPARATE HYDROLYSIS AND FERMENTATION (SHF) AND ENZYMATIC HYDROLYSIS AND

OLEAGINOUS FERMENTATION IN SINGLE REACTOR (SRF) OF STEAM EXPLODED ARUNDO DONAX

54

3.3 MICROALGA-OLEAGINOUS YEAST MIXED CULTURES TO IMPROVE SINGLE CELL OIL

PRODUCTION 64

7

3.4 COMBINED YEAST AND MICROALGA CONSORTIUM CULTIVATION IN PILOT SCALE RACEWAY

POND FOR URBAN WASTEWATER TREATMENT AND POTENTIAL SINGLE CELL OIL PRODUCTION115

4. SUMMARY AND CONCLUSIONS 120

REFERENCES 122

8

LIST OF ABBREVIATIONS

15 FPU & 30 CBU: steam exploded Arundo donax hydrolysate using cellulase from

Trichoderma reesei and β-glucosidase from Aspergillus niger, dose of 15 FPU/g cellulose and

30 CBU/g cellulose, respectively

15 FPU: steam exploded Arundo donax hydrolysate using only cellulase from Trichoderma

reesei, dose of 15 FPU/g cellulose

30 CBU: steam exploded Arundo donax hydrolysate using only β-glucosidase from

Aspergillus niger, dose of 30 CBU/g cellulose

A.U.: absorbance unit

Ac: sodium acetate

ADH: Arundo donax hydrolysate

BBM: Bold Basal Medium

BBM+G: Bold Basal Medium plus D-Glucose

Buffer pH 5.2: hydrolysate of steam exploded Arundo donax using buffer solution at pH 5.2

Buffer pH 6.2: hydrolysate of steam exploded Arundo donax using buffer solution at pH 6.2

CBU: Cellobiase Unit

CFU/mL: Colony Forming Unit/mL

Chl: Chlorella emersonii 312/25

Chla: Chlamydomonas sp.

EFAs: Essential Fatty Acids

FAME: Fatty Acid Methyl Ester

FPU: Filter Paper Unit

G: D-Glucose

G: glucose concentration

G0: initial glucose concentration

GC: Gas Chromatography

Gmax: maximum glucose concentration

HMF: hydroxymethylfurfural

HPLC: High Performance Liquid Chromatography

Lip: Lipomyces starkeyi DBVPG 6193

MP100: medium containing buffer MES hydrate (100 mM)

MES hydrate: 2-(N-Morpholino)ethanesulfonic acid hydrate

9

NaOH pH 5.2: hydrolysate of steam exploded Arundo donax using NaOH solution to adjust

the pH 5.2

PCO2: carbon dioxide fixation rate

PUFAs: Poly Unsaturated Fatty Acids

Px: biomass productivity

SCOs: Single Cell Oils

SHF: Separate Hydrolysis and Fermentation

SRF Buffer pH 5.2: enzymatic hydrolysis and oleaginous Fermentation in Single Reactor

keeping a pH value of 5.2 using a phosphate buffer solutions

SRF NaOH pH 5.2: enzymatic hydrolysis and oleaginous Fermentation in Single Reactor

adding suitable amounts of 1 M NaOH or 0.1 M HCl to maintain a pH value of 5.2 under

discontinuous conditions

SRF without pH control: enzymatic hydrolysis and oleaginous Fermentation in Single

Reactor without pH control

SRF: enzymatic hydrolysis and oleaginous Fermentation in Single Reactor

SSF: Simultaneous Saccharification and Fermentation

UWRP Lip:m.c.: Urban Wastewater in Raceway Pond, using Lipomyces starkeyi and

microalgal consortium ad inoculum

Without enzyme: steam exploded Arundo donax hydrolysate without enzyme dosage

Without pH control: hydrolysate of steam exploded Arundo donax without pH control

WSAELH Chla:Lip 1:1 dil 0, 2, 4: Wheat Straw subjected to Acid Hydrolysis plus

Enzymatic hydrolysis on the Liquid fraction, using Chlamydomonas:Lipomyces starkeyi ratio

of 1:1 and dilution factors of 0, 2, 4

WSAELH: Wheat Straw subjected to Acid hydrolysis plus Enzymatic Hydrolysis on the

Liquid fraction

WSAESH Chla:Lip 1:1 dil 0, 2, 4: Wheat Straw subjected to Acid hydrolysis plus

Enzymatic Hydrolysis on the Solid fraction, using Chlamydomonas:Lipomyces starkeyi ratio

of 1:1 and dilution factors of 0, 2, 4

WSAESH: Wheat Straw subjected to Acid ydrolysis plus Enzymatic Hydrolysis on the Solid

fraction

WSAH Chla:Lip 1:1 dil 0, 2, 4: Wheat Straw subjected to Acid Hydrolysis, using

Chlamydomonas:Lipomyces starkeyi ratio of 1:1 and dilution factors of 0, 2, 4

WSAH: Weat Straw subjected to Acid Hydrolysis

YE: Yeast Extract

10

YEG: Yeast Extract and D-Glucose medium, which contains (g/L): KH2PO4 (3.0), Na2HPO4

(1.0), yeast extract (5.0), glucose (10.0), peptone (5.0)

YGlucose: Glucose Yield

Ylipid: lipid yield

Yx/s: Cell yield coefficient

μG: specific rate for glucose production

μs: specific glucose consumption rate

μx: specific growth rate

𝜈G: glucose productivity

11

1. INTRODUCTION

1.1 A Global Redesign? Ambition of the “circular economy”

In recent years, several steps into the transition toward a biobased economy have been taken.

The ―circular economy‖ has been heralded as one of the most effective instruments for

moving society towards a much needed ‗resource revolution‖, with the aim to provide a better

alternative to the dominant economic development model, so called ―take, make and dispose‖

(Ness, 2008). In the last decade, three major disruptions to the circular economy were

experienced: credit crisis of 2008; Green Fence Operation (GFO), when increased custom

controls at the Chinese borders were implemented, enforcing strict quality criteria to imports

in 2013; the ongoing collapse of crude oil prices. These crisis are major threats to the

established resource recovery systems, but can also serve as an eye opener, forcing us to

reflect upon the necessity for resource recovery (Velis, 2015).

Many adherents of the circular economy approach are strong proponents, on environmental

and ethical premises, of material reuse and recycling (Andersen, 2006). It is rapidly landing

on the world of waste and resources management, becoming a mainstream concept. The

simple and straightforward question in recovering resources, such as biobased materials and

energy, is ―how?‖. In a circular economy, the resource loop would be closed, thus enhancing

its reuse. Other products can be made from plant-based materials that biodegrade into

fertilizer at end of their life. Extending this logic across the economy means a deep change in

the basic structures of industrial systems. In terms of energy, redesigning industry at the

system level would enable efficiency improvements and then potential savings. Part of energy

needed for a circular economy would be provided by renewable sources. Efficient use of

renewable sources for the production of both biobased products and bioenergy should be

driven by well-developed integrated biorefining systems.

1.2 The definition for biorefinery

Biorefinery is the sustainable processing of biomass into a wide spectrum of marketable bio-

products, such as food, feed, biomaterials, and chemicals, and bio-energy, such as fuels,

power, and/or heat (Cherubini, 2010).

The assessment of biorefinery should also take into account the possible consequences related

to the competition for food and biomass resources, the impact on water use and quality,

12

changes in land-use, soil carbon stock balance and fertility, net balance of gas emissions,

impact on biodiversity, potential toxicological risks, and energy efficiency.

The impacts on international and regional dynamics, the end users and consumer needs, and

the investment feasibility are also important aspects that need to be considered. For this

reason the sustainability assessment is not an absolute number, but should be done in

comparison to conventional systems providing the same products and services (de Jong and

Jungmeier, 2015). New biorefinery concepts are, however, still mostly in R&D (Research and

Development) phase or in pilot or small-scale demonstration state, and their industrial

application is still far away. It is expected that these new concepts will be implemented in the

market of different countries in the medium term (2015–2025) (Wagemann et al., 2012)

although current economic conditions (relatively low oil prices, credit crisis, and recessions in

parts of the global economy) might cause severe delays in the market of some biorefinery

concepts.

Many different biorefinery concepts are being developed and implemented. Some of these are

sometimes very complex, they use several feedstocks (e.g., algae, energy crops and wood

chips from short rotation) to coproduce a wide spectrum of bio-products (e.g., bioethanol,

phenols, Single Cell Oils, biodiesel) using technologies that still need to become commercial

in the upcoming years. Among these, particular attention is paid to microbial cultures that can

be established to convert lignocellulosic sugars or low-value hydrophobic substrates into

Single Cell Oils.

1.3 Single Cell Oils

Oils derived from microbial sources are named microbial oils, unicellular oils or single-cell

oils (SCOs). In particular, the term SCOs is used to denote oils of microbial origin. Its

composition is similar to the oil composition of edible plants, animal oils and fats (Kyle and

Ratledge, 1992; Boswell et al., 1996).

The potential of microbial oils and fats has been recognized since the last century. Before the

1980s, several scientists focused their attention on the biochemistry and metabolism

accumulation of lipids in oleaginous microorganisms (Botham and Ratledge, 1978; Botham

and Ratledge, 1979; Gill et al., 1977; Ratledge and Hall, 1977). In the subsequent 20 years,

the biochemical processes and SCO production become one of the most popular research

topic because it was discovered that SCOs can play a critical roles in human health (Sijtsma

and Swaaf, 2004). Subsequently, the microbial screening for the SCO production became the

mission of several scientists (Papanikolaou et al., 2007; Wu et al., 2005). The interest in SCO

13

production continuously increases and it is playing a important role in today's world due to its

potential to solve the current energy crisis. Actually the mechanisms for Single Cell Oil

accumulation by microorganisms are known and this allow to concentrate the researcher focus

on finding low-cost and alternative feedstock and on improving the productivity of SCO

production. It is obvious that SCOs will play a more critical role in the future, and low-cost

substrates, such as lignocellulosic biomasses, will play a key role in the industrialization of

SCO production (Huang et al., 2013). There are several models that describe the SCO

accumulation in oleaginous yeast. Recently, Candrell and Walker (2009) proposed a logistic

model to describe the growth profile; Fakas et al. (2009), using glycerol as a carbon source,

have developed a model for SCO production considering the dependence of nitrogen source.

It is therefore interesting to investigate this aspect partly described by Economou et al. (2010),

matching the experimental data with mathematical model.

1.4 Properties of lignocellulosic materials

Biomass and biomass derived materials have been pointed out to be one of the most

promising alternatives to fossil resources (Keegan et al., 2013; Zhou et al., 2008).

These materials are generated from available atmospheric CO2, water and sunlight through

biological photosynthesis. Therefore, this biomass has been considered to be the only

sustainable source of organic carbon in earth and the perfect equivalent to petroleum for the

production of bio-fuels and bio-fine chemicals with net zero carbon emission. In this context,

lignocellulosic biomass, which is the most abundant and bio-renewable biomass on earth can

play a critical role (Somerville et al., 2010; Zhou et al., 2011). It was estimated that 3.7×109

t

of agricultural residues is produced annually as by-products by agricultural world-wide

industries and that 1376×106

t cellulose and 848×106 t hemicellulose occur globally every

year (Bentsen et al., 2014). Furthermore, it was estimated that one-third of the globally food

produced for human consumption is wasted every year. The overall amount of food wasted

corresponds to around 1.3×109 t (Gustavsson et al., 2011; Gustavsson et al., 2013). These

estimation pointed out that lignocellulosic biomass is the most abundant carbon-neutral

renewable source, which can decrease CO2 emissions and atmospheric pollution. Thus, it is a

promising alternative to limited crude oil, and it has a good potential for the biofuel,

biomolecule and biomaterial production. Furthermore, the major component of lignocellulosic

biomass, the cellulose, is considered the strongest potential candidate for the substitution of

petroleum-based polymers owing to its ecofriendly properties like renewability, bio-

compatibility and biodegradability (Isikgor et al., 2015).

14

Considering the economic point of view, lignocellulosic biomass can be produced quickly and

at lower cost than other agriculturally biofuel feedstocks such as corn starch, soybeans and

sugar cane. It is also significantly cheaper than crude oil (Huber, 2008). On the other hand,

the development of the conversion of lignocellulosic biomass to fine chemicals and polymers

still remains a big challenge for technical and economic obstacles (Himmel et al., 2007; Zhou

et al., 2011) due to biomass recalcitrance that is the result to the terrestrial plant evolution, in

particular the plant cell walls are becoming stronger and more efficient barriers against the

intrusion and degradation from natural infections. Extensive research is currently being

undertaken all over the world to address this problem (Slavin et al., 2011).

Biorefinery and biofuel technologies are developing to refine lignocellulosic biomass for the

production of renewable oil and green monomers (Stöcker, 2008). In addition, the number of

biorefinery-related pilot and demonstration plants has been increasing (Cherubini and

Strømman, 2011). Actually, only few companies, such as Lignol, Verenium and Mascoma,

are focusing their attention on the development of biorefining technologies for the production

of advanced biofuels, biochemicals and biomaterials from non-food cellulosic biomass

feedstocks (Isikgor et al., 2015). Lignocellulosic biomasses are characterized from micro and

macrofibrils organized in crystalline structures. Those tight structures need to be pretreated in

order to make their cellulose, hemicellulose, lignin and small amounts of pectin, protein and

ash accessible to the subsequent hydrolysis step (Galbe and Zacchi, 2002).

The cellulose, hemicellulose and lignin amount is not constant between the lignocellulosic

biomasses. It depends to the species, the age and growing conditions of biomass (Carpita et

al., 2001; Kuhad et al., 1997). The cellulose is the most abundant constituent of plant cell

structure, it is a polysaccharide characterized by a linear sequence of D-glucose molecules

linked by β-1,4 glucosidic bonds with a high degree of polymerisation equal to 10000 or even

higher. This structure results in the formation of intra and intermolecular hydrogen bonds, that

characterize a specific property of crystallinity, even if some sources claim that the structure

could be seen as amorphous (Ding and Himmel, 2006). In addition to being chemically stable

and resistant to microbial degradation (Ward and Moo-Young, 1989), the cellulose fibrils are

responsible for the great tensile strength of the cell wall (Carpita and Gibeaut, 1993). This

structural and inherent integrity of cellulose is believed to play an important role in the

recalcitrance of lignocellulosic biomass (Ding and Himmel, 2008). The hydrophobic surface

involves the formation of water layer that prevents the diffusion of enzymes and degradation

of the products on the surface (Matthews et al., 2006).

15

Hemicellulose is the second most abundant polymer of plant cell structure. Unlike cellulose,

hemicellulose has a random and amorphous structure, which is composed by several

monomers including D-Glucose, D-galactose, D-mannose, D-xylose, L-Arabinose, D-

glucuronic acid and 4-O-methyl-D-glucuronic acid. It is characterized by a degree of

polymerisation lower than 200. Hemicelluloses are imbedded in the plant cell walls to form a

complex network of bonds that provide structural strength by linking cellulose fibers into

microfibrils and cross-linking with lignin.

The lignin is a complex network of phenyl propane units. It is the most important non-

polysaccharide fraction of lignocellulosic biomass. The three characteristic lignin‘s monomers

are: p-cumarilic alcohol, sinapyl alcohol and conferilic alcohol related by different ether

bonds. The lignin also provides protection against chemical and microbial degradation

(Kuhad et al., 1997) [Figure 1.1].

Fig. 1.1: The main components and structure of lignocellulose (SOURCE: Isikgor et al., 2015)

16

1.5 Pretreatment technologies

The lignocellulosic biomass is characterized by complexed structures and requires pre-

treatment processes in order to make those structure accessible to the next step of enzymatic

hydrolysis (Galbe and Zacchi, 2002). The choice of the pretreatment process must also

prevent the carbohydrates degradation because the carbohydrates represent the most

indispensable substrate for the metabolic activity of the microorganisms. The lignocellulosic

biomass pretreatment is also required to avoid the formation of inhibitors directly related to

lignin fraction. Thus, pretreatments are considered one of the most crucial steps since it has a

large impact on all other steps in the conversion process. however, those pretreatment process

are usually energy-expensive steps and significantly affect the cost of the overall process

(Agbor et al., 2011; Sun and Chen, 2002). The pretreatment methodologies can be classified

in physical, chemical-physical, chemical and biological. Some ―physical‖ pre-treatment are

mechanical shredding, pyrolysis and irradiation (McMillan, 1994; Wyman 1996), ―chemical-

physical‖ are steam explosion, AFEX (Ammonia Fiber Explosion), CO2 and SO2 explosion

(Alizadeh et al., 2005; Ballesteros et al., 2000; Dale et al., 1996; Sassner et al., 2005),

―chemical‖ are ozonolysis, acid and basic hydrolysis, oxidative delignification and organosolv

(Arato et al., 2005; Berlin et al., 2006; Karimi et al., 2006a; Karimi et al., 2006b; Schell et

al., 2003) while ―biological‖ pretreatment require fungal and actinomycetes activity (Fan et

al., 1982; Wyman, 1996). Table 1.1 shows the effects of several pre-treatment processes on

the structure of lignocellulosic biomass.

Tab 1.1: Effects on chemical composition and on chemical/physical structure of different pretreatment processes

(H, High; L, Low; ND, Not Determined) (SOURCE: de Jong and Gosselink, 2014)

Pretreatment Sugar

yield

Increases

accessible

surface area

Decrystallizes

cellulose

Removes

hemicellulose

Inhibitor

Formation

Removes

lignin

Alters

lignin

structure

Reuse of

chemicals

Mechanical L

ND

No

Steam explosion H H

H H

L

Liquid hot water H H ND H H

L No

Wet oxidation H or L

ND

No

Dilute acid

H

H

H

Concentrated

acid H H H H H H L Yes

Lime (alkaline)

H ND L L H H Yes

Organosolv H H

H H H L Yes

AFEX/ARP

H H L L H H Yes

Ionic liquid

H H L L H L Yes

Supercritical

fluid

H H H

L

17

It is not possible define the best pre-treatment technology, but it is rather necessary reconcile

the economic and chemical requirements with the industrial implementation (de Jong and

Gosselink, 2014).

1.6 Enzymatic hydrolysis

The pretreatement of lignocellulosic biomasses is important to facilitate access to the

crystalline structure by so-called hydrolytic enzymes, a range of cellulolytic fungi and

bacteria. These microorganisms are being used for a number of industrial purposes, e.g. cotton

processing, detergent enzymes and paper recycling.

Enzymatic hydrolysis of cellulose and hemicellulose is conducted under mild conditions (pH

4.5÷5.0 and 40÷50 °C) which ensure reduced corrosion problems, low energy consumption

and low toxicity of the processed hydrolysates (Taherzadeh and Karimi, 2007).

It is a multi-step and heterogeneous reaction where the insoluble portion of the biomass is

initially broken as a result of synergistic action of different enzymes (Eriksson et al., 2002;

Väljamäe and al., 2003).

These can be divided into the following three types:

Endoglucanases (EG), which hydrolyse internal β-1,4-D glucosidic linkages randomly

in the cellulose chain;

Cellobiohydrolases (CBH, also known as exoglucanases), which progress along the

cellulose line and cleave off cellobiose units from the ends;

β-glucosidases (BG), which hydrolyse cellobiose to glucose and also cleave off

glucose units from cello-oligosaccharides.

The synergistic action of these enzymes is able to increase glucose conversion starting from

lignocellulosic biomass, but is not neglegible the introduction of hemicellulosic enzymes,

such as:

Endo-1,4- β-D-xylanases to make xylan chain;

1,4-β-D-xylosidases to release xylose;

Endo-1,4- β-D-mannanases to break internal bonds to obtain mannans;

1,4-Β-D-mannosidase that cleave molecules of mannooligosaccaridi in mannose.

Cellulases and hemicellulases described above can be produced by bacteria such as

Clostridium, Ruminococcus, Streptomyces, Cellumonas, Bacillus and Ervinia and fungi such

as Trichoderma, Penicillium, Fusarium and Humicola (Rabinovich et al., 2002; Sun and

Cheng, 2002). Among cellulases produced by various microorganisms, those derived from

Trichoderma reesei or from Trichoderma viride have been widely studied and best

18

characterized. It is noted a stability of their enzymatic activity, a resistance to the presence of

inhibitors, but reduced β-glucosidase activity. On the other hand, Aspergillus is able to

overcome this inefficiency by Trichoderma. In many studies it has been observed a synergy

between enzymes derived from two microbial strains to improve the efficiency of the

hydrolysis process (Itoh et al., 2003; Ortega et al., 2001; Tengborg et al., 2001; Wyman,

1996) [Figure 1.2].

Fig. 1.2: Simplified diagram of lignocellulose hydrolysis showing synergism and limiting factors. Cellulose is symbolized

straight lines. 1: Product inhibition of β-glucosidase (BG) and cellobiohydrolase (CBH) by glucose and cellobiose. 2: CBH

hydrolysing from the end of a cellulose chain. 3 and 4: Hemicelluloses and lignin associated with or covering the microfibrils

prevent the cellulases from accessing the cellulose surface. 5: Enzymes (both cellulases and hemicellulases) can be unspecifically adsorbed onto lignin particles or surfaces. 6: Denaturation or loss of enzyme activity due to mechanical shear,

proteolytic activity or low thermostability (SOURCE: Jørgensen et al., 2007).

The enzymatic hydrolysis presents several obstacles that prevent the implementation on an

industrial scale, the first is associated to enzyme cost but in recent years is much less incident

for the intensification of research by, for example, Novozymes and Genecor aimed to reduce

the use of these enzymes in order to minimize process operating costs (Zhang et al., 2006).

Other parameters, already mentioned, such as pH, temperature and residence time of the

process linked to other not secondary aspects such as substrate concentration which helps to

19

increase inhibition phenomena. The Equation (1.1) that allows to describe the inhibition

phenomenon is associable to Michaelis-Menten equation:

v =

(

)

(1.1)

is assumed constant, is the value of catalytic activity, represents the initial

concentration of enzyme, S is substrate concentration, is Michaelis-Menten constant, I

inhibitor concentration and represents the dissociation constant of enzyme. In case of

product inhibition (β-glucosidase inhibition due to the presence of an excessive glucose

concentration), the term

becomes

(P is the product and the dissociation constant). If

P increases due to the reaction catalyzed by enzyme, also the

increases causing a decrease

of reaction rate, strictly related to the increase of the product concentration.

Since industrial processes typically require high product concentrations, above mentioned

inhibition phenomena significantly reduce reaction rate and the efficiency of enzymatic

hydrolysis reactions in batch and continuous processes (Andric et al., 2010). One of

actionable solutions to overcome this problem is to increase the ratio enzyme/substrate,

however, contrary to the fulfillment of the objective to minimize operating costs or to add

surfactants to change surface properties of cellulase, or to recover the enzyme fraction, once

immobilized, through recycling mechanisms (Tu et al., 2006). All of these techniques for

recycling and reducing enzyme adsorption have so far only been tested at laboratory scale.

Furthermore, most of the studies do not include cost calculations to evaluate the feasibility of

addition of different compounds to reduce enzyme binding or the costs of recycling.

Therefore, the ability to scale up the techniques, the robustness and feasibility still needs to be

demonstrated (Jørgensen et al., 2007).

Finally the development of methodologies that include the progressive glucose consumption

produced during the hydrolysis step. This type of process is commonly called Simultaneous

Saccharification and Fermentation (SSF) where hydrolysis and fermentation are located in a

single reactor. This process has been considered as preferably process because of reduced

operation all costs, lower enzyme requirement and increased productivity (Alfani et al., 2000).

The drawback, still under study, is the ability to reconcile the cultivation requirements of

yeasts that provide temperatures of 30÷37 °C and enzyme activity that is high at 40÷50 °C

(Olsson et al., 2006; Saha et al., 2005; Wingren et al., 2003). SSF from lignocellulosic

20

biomass is commonly studied for bioethanol production (Kim et al., 2008; Lynd et al., 2008),

however is still unexplored the use of SSF for single cell oil production (Liu et al., 2012). The

reason is attributable to the different metabolic activity of oleaginous yeasts that produce

lipids in intracellular spaces of solid phase, and to the difficulty of oxygen supply due to the

increase of culture medium viscosity especially in case of high concentrations of processed

biomass (Ageitos et al., 2011; Beopulos et al., 2011; Liu et al., 2012).

1.7 Oleaginous microorganisms

1.7.1 Oleaginous yeasts

The growing importance of Single Cell Oil applications in nutraceutical and pharmaceutical

field, and the possibility to use microbial oils also for biodiesel production has created

scientific appeal aimed to obtain alternative ways such as offered by oleaginous yeasts.

Oleaginous yeasts are not influenced by climate variability, they grow in presence of different

substrate sources as hexose and pentose sugars with high growth rates (Chi et al., 2010).

Generally, they have the ability to accumulate Single Cell Oils more than 20-25% of their

weight (Ratledge and Wynn, 2002; Beopoulos and Nicaud, 2012) but not more than 65% of

their dry weight in specific growth conditions (Angerbauer et al., 2008).

Oleaginous yeasts are attracting an increasing interest for their growth requirements that are

the effect of poor request of nitrogen source and the ability to trigger a cascade of reactions

leading to intermediate compounds formation such as Acetyl-CoA (Li et al., 2007;

Papanikolaou and Aggelis, 2011) and accumulation mechanisms (Nagamuna et al., 1985) that

are derived from tricarboxylic acid cycle (TCA cycle) that takes place in mitochondria of

eukaryotic microorganisms [Figure 1.3].

Fig. 1.3: Citrate/malate cycle and the transhydrogenase cycle (TCA) are acetyl-CoA and NADPH precursors for lipogenesis

in oleaginous microorganisms (SOURCE: Ratledge, 2004)

21

The eukaryotic microorganisms, oleaginous and not oleaginous, share the same biosynthetic

pathway, however, exists a fundamental difference that distinguishes them in the presence of

a carbon source excess. In a medium with abundant carbon source and limiting amounts of

nitrogen, when all the nitrogen source has been consumed, oleaginous microorganisms utilise

carbon source for lipid synthesis resulting in an excess of triacylglycerols; while not

oleaginous microorganisms convert carbon source in polysaccharides (glycogen, glucans,

mannans, etc.) and are not predisposed to accumulate lipids (maximum up to 10-20%). As

above mentioned, since the biosynthetic pathway of fatty acids is the same, in oleaginous

microorganism, the reason is related to the production of acetyl-CoA in the cytosol as a

precursor for FAS (Fatty Acids Synthetase) and to the production of NADPH, which is used

as a reducing agent in the synthesis of fatty acids (Ratledge, 2004).

Fatty acid biosynthesis in almost all organisms, is directed to C16 and C18 saturated fatty

production that are modified by a sequence of enzymes (elongase and desaturase), able to

insert progressively a defined range of unsaturations. In oleaginous yeasts, most fatty acids

are commonly represented by oleic acid (C18:1, n-9), linoleic acid (C18:2, n-6), palmitic acid

(C16:0) and palmitoleic acid (C16:1), as well as by C18:3 or alpha-linolenic acid that, in

general, represent less than 10% of the total (Xu et al., 2012).

It has been proven in recent years that yeasts such as Rhodotorula glutinis (Easterling et al.,

2009), Rhorosporidium toruloides (Li et al., 2007), Trichosporon fermentans (Zhu et al.,

2008) and Lipomyces starkeyi (Angerbauer et al., 2008; Zhao et al., 2008) have a potential

production also because they can be cultured in simpler media made with low cost substrates

(Huang et al., 2013). A further important advantage to use such oleaginous yeasts comes from

the ability to produce lipids under aerobic conditions from residual substances, without

adding expensive nutrients [Table 1].

22

Tab 1.2: Oleaginous yeast species for single-cell oil production from cellulosic sugars (SOURCE: Abghari and Chen, 2014)

Oleaginous yeast

species Substrate Comments Reference

Cryptococcus sp. Glucose and corncob hydrolysate

Batch and fed-batch to reach lipid

content and productivity of more

than 61% w/w and 1 g/L/day,

respectively

Chang et

al., 2013

Cryptococcus

curvatus Oligocelluloses and oligoxyloses

Batch to reach lipid content and

coefficient of more than 30% and

0.17 g/g sugar, respectively

Gong et al.,

2014

Lipomyces

starkeyi Corncob acid hydrolysate

Batch to reach lipid content and

yield of 47% and 8.1 g/L,

respectively

Huang et

al., 2014

Lipomyces

starkeyi

Co-fermentation of cellobiose and

xylose

Batch to reach lipid content and

coefficient of 55% and 0.19 g/g

sugar, respectively

Gong et al.,

2012

Rhodosporidium

toruloides Jerusalem artichoke

Batch and fed-batch to reach lipid

content and yield of 43.3-56.6%

and 17.2-39.6 g/L, respectively

Zhao et al.,

2010

Rhodotorula

graminis

Undetoxified corn stover

hydrolysate

Batch to reach lipid productivity

and lipid content of 0.21 g/L/h and

34% w/w, respectively

Galafassi et

al., 2012

Trichosporon

coremiiforme Corncob acid hydrolysate

Batch to reach lipid content and

yield of 37-40% and 7.7-9.8 g/L,

respectively

Huang et

al., 2013b

Trichosporon

cutaneum

Diluted acid pretreated and

biodetoxified corn stover

Simultaneous saccharification and

fermentation to reach lipid yield of

3.03-3.23 g/L

Liu et al.,

2012

Yarrowia

lipolytica

Sugarcane bagasse and rice bran

hydrolysate

Batch to reach lipid content and

yield of 48-58.5% and 5.16-6.68

g/L, respectively

Tsigie et al.,

2012

The main lipid molecules present in yeasts are triacylglycerols (TAGs) and steril-esters (SE)

and are accumulated in the cells during stationary growth phase. TAGs have a distribution of

the acyl substituents very similar to the plant oils, in particular the central position is occupied

almost exclusively by an unsaturated acyl group. Since these molecules are without positive

or negative charge, they can‘t be part of cell membranes, however, they are sequestered in

hydrophobic particles called lipid particles (LP) or lipid bodies. The lipids are accumulated in

the form of micro droplets, and composed almost entirely of triacylglycerols. Their extraction

implies other lipid fractions (phospholipids, sterols, steril-esters and other) associated to the

cell membrane and the presence of free fatty acids due to and uncontrolled lipolysis which

occurs during extraction process (Cordisco, 2009).

Lipid accumulation implies three consecutive stages: synthesis, storage and mobilization.

These processes require the interaction of various cellular components, in particular the

plasmic membrane, the lipid particles (LP) and the endoplasmic reticulum (ER) (Cordisco,

2009). The LP biogenesis is still not completely known, but one of the most accepted models

is so-called ―budding model‖, which assumes the LP formation from the endoplasmic

23

reticulum membrane [Figure 1.4]. According to this model, specific enzymes synthesize

neutral lipids, such as TAG and SE, which accumulate in the phospholipid bilayer of ER

membrane. This accumulation continues until it comes off a mature LP (Ratledge, 2004;

Czabany et al., 2007; Courchesne et al., 2009).

Fig. 1.4: LP formation from endoplasmic reticulum (ER) trough budding model (SOURCE: Ratledge, 2004)

1.7.2 Microalgae

Microalgae are not the newly known microorganisms for human beings, since they are widely

used for decades as the feedstock for traditional applications in cosmetic, pharmaceutical and

nutrition sectors. In addition, a variety of bio-active substances, such as carotenoids,

polysaccharides and β-carotene, can be produced. The forms of microalgal products include

tablets, capsules, liquids, pure molecules with high values such as fatty acids, pigments and

stable isotope biochemicals, and cosmetics found in face and skin care products, such as anti-

aging creams, refreshing or regenerant care products, emollient and anti-irritant in peelers

(Koller et al., 2012; Samarakoon et al., 2012).

Microalgae are prokaryotic or eukaryotic photosynthesis microorganisms that can grow

rapidly and live in harsh conditions due to their unicellular or simple multicellular structure.

Examples of prokaryotic microorganisms are Cyanobacteria (Cyanophyceae) and eukaryotic

microalgae are for example green algae (Chlorophyta) and diatoms (Bacillariophyta) (Li et

al., 2008a; Li et al., 2008b). In particular, microalgae may assume many types of

metabolisms, such as photoautotrophic, heterotrophic, mixotrophic and photoheterotrophic

(Mata et al., 2010; Zhao et al., 2012), offering the highest yields of lipids, mainly made of

triglycerides (Chen et al., 2009). These are composed of saturated and unsaturated fatty acids

with 12–22 carbon atoms, some of them of 𝜔-3 and 𝜔-6 families. At present, the research has

24

focused on the photoautotrophic production of microalgal oil using light as energy source, but

there are significant drawbacks associated with photoautotrophic algal cultures for Single Cell

Oil production. First, it is difficult to solve the contradiction between accumulation of

biomass and lipid synthesis during the microalgal life cycle (Liu et al., 2011). Second, light

attenuation is unavoidable for photoautotrophic cultures from lab to pilot scale, and it may

significantly reduce productivity (Wilhelm and Jakob, 2011).

Many microalgae strains such as Chlorella sorokiniana (Chen et al., 2013), Chlorella

saccarophila (Herrera-Valencia et al., 2011), Nannochloropsis sp. (Jiang et al., 2011;

Wahidin et al., 2013; Chiu et al., 2009), Neochloris oleoabundans (Li et al., 2008c),

Cladophora fracta (Demirbas, 2009), Chlorella protothecoides (Demirbas, 2009; Xiong et

al., 2008), Chlorella vulgaris (Liu et al., 2008) and Chlamydomonas sp. (Ho et al., 2014;

Nakanishi et al., 2014) could supply Single Cell Oil production.

As above explained, microalgae cultures can have different types of metabolisms that are

distinguished in autotrophic, heterotrophic, mixotrophic and phototrophic. In autotrophic

condition, microorganisms obtain energy absorbing light energy for CO2 reduction and they

imply O2 release. Most of algae belong to this category and require minimal amount of

organic compounds for growth, such as vitamins. Heterotrophy it means that are required

organic carbon sources to support the growth. Some microalgae are able to assimilate organic

carbon as a source of energy to grow in dark mode. Heterotrophic growth solves technical and

physiological problems related to the presence and distribution of light and CO2 associated to

autotrophic growth. Therefore, it offers the possibility of increasing cell concentration and

productivity. However, mixotrophic growth is defined as a growth, where CO2 and organic

carbon source are assimilated simultaneously, using photosynthetic metabolism and cell

respiration. Normally, cell growth rate in mixotrophy is approximately the sum of growth rate

in heterotrophic and autotrophic mode, and the advantages of heterotrophic growth, such as

high concentration and productivity, are also applicable to mixotrophic growth.

Finally, photoheterotrophy also known as photo assimilation or photo metabolism, describes a

metabolism in which the presence of light is required to use organic compounds as a carbon

source.

Not only the carbon (in form of CO2 and organic source) is required for the microalgae

metabolism, but also vitamins, salts, other nutrients (nitrogen and phosphorus) and the

balance of following parameters such as oxygen, carbon dioxide, pH, temperature, and light

intensity (Chojnacka et al., 2004).

25

The pathway of triglyceride synthesis in microalgae consists of three steps (Huang et al.,

2010):

1) formation of acetyl-CoA in the cytoplasm (acetyl-CoA is the initiator, its formation takes

place in the chloroplast, where it is formed an intermediate, glyceraldehyde phosphate (GAP),

that is transferred to the cytoplasm and subsequently consumed. After the export of GAP from

chloroplast to cytoplasm, the carbon source is directed to the sugar synthesis (that represent

the main storage products in the cytoplasm of plant cells) or oxidation through the glycolytic

pathway to pyruvate. Therefore, a part of the exogenous glucose is directly converted into

starch and the rest is oxidized via glycolysis);

2) elongation and desaturation of carbon chains of fatty acids (elongation depends mainly on

the reaction of two enzyme systems that include acetyl-CoA and requires the collaboration of

malonyl-CoA [Figure 1.5]. Desaturation of carbon chain occurs further elongation of carbon

chain takes place to produce long-chain fatty acids which are unusual in normal plant oils);

Fig. 1.5: Fatty acid synthesis (SOURCE: Shen and Wang, 1989)

3) triglyceride biosynthesis of triglycerides (generally, L-𝛼-phosphoglycerol and acetyl-coA are two major primers. The reaction steps are shown in Figure 1.6);

26

Fig. 1.6: Triglyceride biosynthesis (SOURCE: Huang et al., 2010)

1.8 Cultivation strategies

The efficiency of the cultivation systems depends on several factors, including the

minimization of operating and plant costs.

As concerns oleaginous fermentations from lignocellulosic biomasses to SCO production,

there are different processes which can be designed in several ways. The common is to

separate hydrolysis and fermentation step (Separate Hydrolysis and Fermentation, SHF). The

major advantage of SHF method is that is possible to carry out hydrolysis (40÷50 °C) and

fermentation (30÷37 °C) at their own optimum conditions. The main drawback of SHF is the

inhibition of hydrolytic enzymes activity by released sugars. A new perspective is offered by

combining enzymatic hydrolysis of pretreated lignocelluloses and fermentation in a single

step or in a single reactor. A further advantage is that the sugars produced by hydrolysis are

immediately consumed by fermenting microorganisms, and possible substrate inhibition of

the enzymes during the hydrolysis is limited (Wingren et al., 2003). This process has been

mainly studied for bioethanol production. The major advantage compare to SHF is that the

released glucose is immediately consumed by fermenting microorganisms and the reduction

in material cost since only one reactor is needed. A possible limitation to the applicability of

the processes in single reactor is the different requirements of hydrolytic enzymes and

oleaginous yeasts in terms of operating temperature and pH. Therefore, for the industrial

scale, new improvements in enzyme technology (e.g. thermostable cellulases and higher

inhibitor tolerance) are required (Viikari et al., 2007). Only few papers have been devoted to

the enzymatic hydrolysis coupled with oleaginous yeast growth in a single reactor. Liu et al.

27

(2012) have studied integration of lipid production and enzymatic hydrolysis of corn stover,

using diluted acid pretreated. The lipid concentration obtained were 3.03 g/L and 3.23 g/L

using Trichosporon cutaneum yeast cultivated in 5 L and 50 L stirred-tank reactor

respectively. Gong et al. (2013) have described Simultaneous Saccharification and enhanced

lipid production of glucose and cellulose by Criptococcus curvatus. When cellulose was

loaded at 32.3 g/L, lipid yield reached 0.20 g/g of cellulose. Gong et al. (2013) have

demonstrated that Criptococcus curvatus can be utilize either oligocelluloses or oligoxyloses

as the sole carbon source for microbial oil production.

An interesting perspective, yet not adequately deepened by the research community, is

represented by mixed cultures of microorganisms common in natural ecological system.

When using a mixed culture, two or more microorganisms are synchronously cultivated

within the same medium, where these microorganisms can mutually exploit complementary

metabolic activities to survive, grow and reproduce (O’Reilly and Scott, 1995).

In the literature review little information on Single Cell Oil production is available regarding

microalga and oleaginous yeast mixed cultures. Many methods and techniques, such as the

use of bioreactors, the use of low-cost carbon substrates like industrial wastes or hydrolysates

of lignocellulosic biomasses have been developed to reduce the cost of SCO production and identify a convenient solution to achieve those targets.

28

AIM OF THE THESIS

The research activity has been developed according to the general plan and, in particular, to

the respect of the goal that was to develop a new approach in biorefinery focused on new

cultivation strategies. The initial study was, therefore, concentrated in particular on the

optimization of different process parameters (temperature, pH, enzyme dosage) related to the

enzymatic hydrolysis, with the aim of assessing the applicability in enzymatic hydrolysis and

oleaginous fermentation in a single reactor (SRF). During the experimental activity, it has

been investigated the influences of environmental factors as initial pH, but also inhibitory

effects of organic acids, furans from hexoses decomposition and phenols from lignin

decomposition, released during hydrolysis and fermentation. Thus, this new process has

offered an opportunity for an integrated process to reduce time and costs for Single Cell Oil

production from lignocellulosic biomasses.

One of the objectives was also to investigate the symbiotic interactions of yeast and microalga

mixed cultures to confirm that both microorganisms could significantly enhance biomass and

lipid production. In theory, there are synergistic effects on gas, substance exchange and pH

adjustment in the mixed culture system of oleaginous yeasts and microalgae based on the

mutually beneficial relationship. In yeast-microalga mixed cultures, microalgae could act as

an O2 generator for the yeast while the yeast provided CO2 to microalgae and both carried out

production of lipids. It was also evaluated using lignocellulosic hydrolysates. Finally, the

combined growth of yeast and microalgae using urban wastewater as substrate was studied

and the lipid biomass concentration during the culture growth phases was monitored. Most of

the experimentations on the combined yeast and microalgal cultures reported in literature

remained in little scale or in laboratory controlled conditions, while in this work the

cultivation was conducted in a 200 L raceway pond operating outdoor.

29

2. MATERIALS AND METHODS

2.1 Pretreatment of lignocellulosic material

Arundo donax (giant reed) is a perennial, rhizomatous grass, classified as energy crop

(Angelini et al., 2005), with different characteristics such as high dry biomass yield (30÷40

ton ha-1

year-1

) (Angelini et al., 2009; Mantineo et al., 2009), the ability to grow in marginal

lands (Nassi et al., 2011), with reduced input of water (Lewandoski et al., 2003), a high

content in cellulose and hemicellulose (about 60%) (Komolwanich et al., 2014; Shatalov and

Pereira, 2013) that make this interesting crop for bioethanol, biodiesel and bio-polymer

production (Williams and Biswas, 2010; Pirozzi et al. , 2010).

It has been pretreated in a continuous pilot plant for the steam explosion (mod. StakeTech

System Digester) located at ENEA–Trisaia Research Centre (Rotondella, Matera, Italy).

Steam explosion of biomass is a pretreatment process that opens up the fibers, and makes the

biomass polymers more accessible for subsequent processes, i.e. fermentation and hydrolysis.

It is considered to be one of the most important. Its attractive features, in comparison to

autohydrolysis, pulping, and other methods, include the potential for significantly reducing

the environmental impact, the investment costs, and the energy consumption (Avellar and

Glasser, 1998).

The biomass has been steam exploded, processing 150-200 kg/h of dry biomass, to which

water was added to raise the intrinsic humidity up to 50%. The pretreatment was carried out at

210 °C for 4 minutes. The severity factor (SF) was determined to be 3.84 according to the

following Equation 2.1 (Garrote et al., 1999):

S log (R) log (t eT-1

14.75) (2.1)

where t is pretreatment time and T is pretreatment temperature.

2.2 The hydrolysis of lignocellulosic biomass

2.2.1 Arundo donax hydrolysate

Steam exploded Arundo donax was mixed with distilled water at pH 5.2, to obtain a solution

with 5% w/v solid content and it was treated with commercial enzymes purchased from

Sigma-Aldrich consisting of cellulase from Trichoderma reesei ATCC 26921 and β-

glucosidase from Aspergillus niger.

30

Enzyme complexes used for the biomass pre-treatment usually consist of 1,4-β-D-

glucanohydrolases (endoglucanases), 1,4-β-D-glucan cellobiohydrolases (exoglucanases), and

β-D-glucoside glucohydrolase (β-glucosidase or cellobiase). A mix of two enzymes with

cellulolytic activity was used to reduce the operating costs. The reason was also related to the

choice of steam explosion as pretreatment because is able to separate hemicellulose from

cellulose and lignin.

Cellulase activity was measured following the NREL filter paper assay (Adney and Baker,

1996; Ghose 1987) and reported in Filter Paper Units per milliliter of solution (FPU/mL). β-

glucosidase activity was measured using the method described by Wood & Bhat (1988) and

reported in cellobiase units (CBU). Enzymatic hydrolysis was performed in flask with 200

mL of slurry. Different hydrolysis tests were carried out to explore different process

conditions. Initially, only cellulase from Trichoderma reesei ATCC 26921 (15 FPU/g of

cellulose) was introduced as control. Tests were carried out also to evaluate synergistic and

inhibition phenomena, using cellulase and β-glucosidase with different dosages. As just

mentioned, it was decided to introduce cellulase from Trichoderma reesei ATCC 26921 (15

FPU/g of cellulose) and β-glucosidase from Aspergillus niger (30 CBU/g of cellulose) (Gong

et al., 2013). To evaluate combined effects, fermentable sugar conversion and enzymatic

hydrolysis yield, cellulase and β-glucosidase were introduced with a concentration equal to

7.5, 15, 30 FPU/g of cellulose and 15, 30, 60 CBU/g of cellulose, respectively.

Tests were performed to evaluate inhibition effects, introducing progressively a higher β-

glucosidase concentration 30, 60, 90, 120, 180 CBU/g of cellulose combined with a

concentration of cellulase of 15 FPU/g of cellulose.

The tests were conducted initially in batch mode with 200 mL of slurry (5% w/v of steam

exploded Arundo donax), at 50 °C, initial pH 5.0, for 72 hours with agitation speed of 160

rpm (MINITRON, Infors HT, Switzerland).

The temperature has an important effect on the enzymatic activity. In fact, denaturation

phenomena are caused by the decline of natural enzyme configuration above 60 °C

(Parameswaran et al., 2011).

Tests were carried out to combine process conditions of enzymatic hydrolysis with oleaginous

fermentation in terms of temperature and pH, with the aim to perform the tests in a single

stage or in a single reactor. The selected temperatures were 30 °C and 40 °C in addition to 50

°C, already explored, and with optimal conditions obtained in previous tests.

Finally, steam exploded Arundo donax was suspended in different media in terms of

composition and pH. Subsequent tests were performed under following conditions:

31

• without pH control (without pH control);

• with phosphate buffer ( . 5 M) at pH 5.2 (Buffer pH 5.2);

• with phosphate buffer ( .2 M) at pH 6.2 (Buffer pH 6.2);

• with NaOH solution to pH 5.2 (NaOH pH 5.2).

The temperature was 40 °C, the speed rotation 160 rpm, the time 72 hours and the chosen

optimum enzyme dosage as obtained from previous tests. The pH was monitored and adjusted

using NaOH and HCl in the sample NaOH pH 5.2. All the tests were carried out, at least, in

duplicate.

2.2.2 Wheat straw hydrolysate

Hydrolysis was also conducted on no pretreated wheat straw in powder. Acid hydrolysis was

performed, using the following parameters: H2SO4 1% v/v, 121 °C and 21 min.

The acid hydrolysate was then subjected to enzymatic hydrolysis, treating the solid fraction

and the liquid fraction using: cellulase from Trichoderma longibrachiatum (1% w/w),

xylanase from Trichoderma longibrachiatum (1% w/w), 50 °C, 72 h, and MES buffer 100

mM and NaOH 0.1 M, where required to bring the pH to a value equal to 5.0. Hydrolysates

were employed for the next stage of fermentation after vacuum pump filtration.

The samples were classified as:

- Wheat Straw subjected to Acid Hydrolysis (WSAH);

- Wheat Straw subjected to Acid Hydrolysis plus Enzymatic hydrolysis on the Liquid

fraction (WSAELH);

- Wheat Straw subjected to Acid Hydrolysis plus Enzymatic hydrolysis on the Solid

fraction (WSAESH).

2.3 Separate hydrolysis and fermentation (SHF) and enzymatic hydrolysis and

oleaginous fermentation in single reactor (SRF) of steam exploded Arundo donax

After the optimization of process parameters related to enzymatic hydrolysis, the subsequent

stage of oleaginous fermentation was considered.

Lipomyces Starkeyi DBVPG 6193 was used as oleaginous yeast, purchased from the Culture

Collection of Dipartimento di Biologia Vegetale (University of Perugia, Italy). The strain was

maintained at 5 °C on a YPD solid medium with the following composition (g/L): yeast

extract (10), peptone (20), D-glucose (20), agar (20). Prior to fermentation, yeast was grown

in a 100 mL Erlenmeyer flasks with an initial volume of 50 mL which contained (g/L):

32

KH2PO4 (3.0), Na2HPO4 (1.0), yeast extract (5.0), glucose (10.0), peptone (5.0). The pH was

adjusted to 5.5÷6 and, prior to inoculation, the pre-culture broth was sterilized at 121 °C for

21 min. Culture media for Single Cell Oil production were inoculated with 5% v/v of pre-

culture media. The incubation of the preculture media were carried out at 30 °C, 160 rpm for

48 hours (Minitron HT Infors, Switzerland) and microbial biomass analyzed through

turbidometry at 600 nm up to 1 A.U. (Absorbance Unit) and then used as inoculum. Tests in

Separate Hydrolysis and Fermentation (SHF) and enzymatic hydrolysis and Fermentation in

Single Reactor (SRF) were conducted using 500 mL aerobic flasks.

SRF experiments were performed under non-sterile conditions to preserve protein and

carbohydrate integrity, preventing the Maillard reaction (Rosenthal et al., 1996) as well to

retain the most likely parameters for industrial applications. Suspension of steam-exploded

Arundo donax (5% w/v) was used in an Erlenmeyer flask with a working volume of 200 mL.

A commercial mixture containing 15 FPU/g of cellulose of Cellulase from Trichoderma

reesei and 30 CBU/g of cellulose of Cellobiase from Aspergillus niger (Sigma-Aldrich) were

employed. After 48 h, a suspension of yeast cells (Lipomyces Starkeyi) was added to the same

flask using 5% v/v concentration.

The pH control was carried out during the experimental tests to limit the pH fluctuations. The

enzymatic hydrolysis of the lignocellulosic materials usually leads to a pH reduction, whereas

the oleaginous fermentation causes a pH increase.

Consequently, the experimental tests in enzymatic hydrolysis and Fermentation in Single

Reactor (SRF) were performed following three different protocols as regards pH control:

(a) without pH control (SRF without pH control),

(b) keeping a pH value of 5.2 using a phosphate buffer solutions (SRF Buffer pH 5.2);

(c) adding suitable amounts of 1 M NaOH or 0.1 M HCl to maintain a pH value of 5.2 under

discontinuous conditions (SRF NaOH pH 5.2).

The test in Separate Hydrolysis and Fermentation (SHF) were performed in the same culture

conditions observed for the tests just described, except the separation of hydrolysis and

fermentation steps. The highest pH values (5÷6) were observed when adopting a

semicontinous control (c). In the presence of a phosphate buffer solution (b), the measured pH

was in the range 4.5-5.0. When no pH control was adopted (a), the lowest values of pH were

observed, as the pH of the medium dropped below 4.

In a typical SRF test, all the enzymes were added at the beginning and the biomass was

hydrolysed at 40 °C and 160 rpm (Minitron, Infors HT, Switzerland). After 48 hours,

temperature and stirring rate were changed to 30 °C and 160 rpm, respectively, and yeast cells

33

(Lipomyces Starkeyi DBVPG 6193) were inoculated. All the tests were carried out, at least, in

duplicate.

2.4 Mixed oleaginous yeast-microalga cultures

In an attempt to improve lipid yields, it was evaluated synergistic effect of microalga-

oleaginous yeast mixed cultures. The tests were conducted, initially, with the aim to find

optimum culture conditions for microalga strains. The seed culture was pre-cultivated onto

100 mL of BBM medium at in an incubator shaker at a shaking speed and continuous

illuminated by 150 µmol m-2

s-1

cool-white fluorescent lamps.

Chlorella emersonii 316/25, Chlorella protothecoides 165, Chlorella saccarophila 042,

Viridiella fridericiana 035, Chlorella zofingiensis 252, Chlorella sorokiniana 317 were

achieved from the ACUF collection of the Department of Biological Science at the University

of Naples, Federico II (http://www.biologiavegetale.unina.it/acuf.html). These algae have

only minimal requirements to the medium for the growth. There is the need of inorganic ions

and a minimal quantity of organic compounds, such as vitamins. Carbon, nitrogen and

phosphorus are the most important nutrients for the autotrophic algal growth (Becker et al.,

1986). Bold Basal medium (Bold, 1949) is an artificial freshwater medium, which is practical

for growing green algae. Bold Basal Medium (BBM) supplemented with NaNO3 (40 mg/L) as

nitrogen source was adopted. The medium was autoclaved for 20 minutes. The final pH

should be 6.8. The BBM medium contained the following components: CaCl2·2H2O (1.7 ·1 -

4 M), KH2PO4 (1.29·1

-3 M), EDTA anhydrous (1.71·1

-4 M), KOH (5.52·1

-4 M), K2HPO4

(4.31·1 -4

M), NaCl (4.28·1 -4

M), MgSO4·7H2O (3. 4·1 -4

M), NaNO3 (2.94·1 -3

M), H3BO3

(1.85·1 -4

M), FeSO4·7H2O (1.79·1 -5

M), H2SO4 (1.79·1 -5

M), ZnO4·7H2O (3. 7·1 -5

M),

MnCl2·4H2O (3. 7·1 -5

M), MnO3 (4.93·1 -6

M), CuSO4·5H2O (6.29·1 -6

M),

Co(NO3)2·6H2O (1.68·1 -6

M). The concentration of inoculum was equal to 10% v/v.

The flasks were incubated at 30 °C and under light intensity around 100 μmol photons m-2

s-1

.

All chemicals were purchased from Sigma-Aldrich. Chlorella emersonii 316/25 and

Lipomyces starkeyi DBVPG 6193 were incubated from BBM and YEG agar slant,

respectively, to 500 mL Erlenmeyer flasks containing 100 mL seed media. The culture flasks

were inoculated, separately and simultaneously, to achieve the initial cell density of 5.5·105

Cells/mL for both microorganisms. The cultures performed containing both inorganic and

organic substrates and were grown in:

- BBM medium (BBM);

- BBM plus D-Glucose (10 g/L) (BBM+G);

http://www.biologiavegetale.unina.it/acuf.html

34

- Yeast Extract and D-Glucose medium, which contains (g/L): KH2PO4 (3.0), Na2HPO4

(1.0), yeast extract (5.0), glucose (10.0), peptone (5.0) (YEG).

The pH of media was adjusted to 6 and prior to inoculation, the culture broth were sterilized

at 121 °C for 21 min.

In the last step steam exploded Arundo donax was used. Arundo donax hydrolysate (ADH)

was inoculated using Lipomyces starkeyi DBVPG 6193 and microalga strains, Chlorella

emersonii 316/25, simultaneously. The initial cell density was 5.5·1 5 Cells/mL for both

microorganims. All the tests were carried out, at least, in duplicate with the same operating

conditions as above mentioned.

For the experimental activity performed at INRA-LBE (Laboratoire de Biotechnologie de

l'Environnement) in Narbonne (France), Chlamydomonas sp. was used. It is a model green

alga (Harris, 2001) that can grow mixotrophically, using various carbon sources and then it is

studied for its behavior in different operating conditions. For example it can uptake acetate,

which is then incorporated into citric cycle (Johnson & Alric, 2012).

The aim of this experimental activity was to verify the differences of kinetic parameters

related to different growth conditions. The cultures were performed in 100 mL of MP or

enriched MP medium with glucose, yeast extract and sodium acetate, added using the

concentrations as explained subsequently. All the tests were carried out at 24 °C in an

incubator shaker and continuous illuminated by 100 µmol m-2

s-1

cool-white fluorescent

lamps. The MP100 medium contained the following components (per liter):

Buffer MES hydrate (100 mM), Beijerinck solution 40x (25 mL), KPO4 (1 M), Hutner‘s trace

elements (1 mL).

After the autoclaving, the pH was set to 6.25, as obtained from previously optimization (data

not shown). Tests were conducted using enriched MP100 medium, adding D-Glucose (G) (10

g/L) and Yeast extract (YE) (0.5 g/L) inoculated by Chlamydomonas sp. (Chla), Lipomyces

starkeyi DBVPG 6193 (Lip) and both strains (Chla:Lip):

- MP100+G+YE Chla;

- MP100+G+YE Lip;

- MP100+G+YE Chla:Lip 1:1, 2:1, 3:1.

Preliminary tests were carried out to evaluate the best Chlamydomonas sp./Lipomyces starkeyi

DBVPG 6193 ratio. The cultures were inoculated by 5% v/v of the seed precultures. The same

operating conditions were used for enriched MP100 medium adding also sodium acetate (Ac)

(1.5 g/L). The samples were marked as:

35

- MP100+G+YE+Ac Chla;

- MP100+G+YE+Ac Lip;

- MP100+G+YE+Ac Chla:Lip 1:1, 2:1, 3:1.

Sodium acetate was added to verify the ability of yeast and microalga to metabolise an

alternative carbon source represented by the organic acid and to simulate realistic growth

conditions in presence of lignocellulosic biomass hydrolysates. All the tests were carried out,

at least, in duplicate.

In order to verify the growth results in mixed cultures, two different vials were performed,

where O2 was replaced by N2 and marked as:

- MP100+G+YE Chla:Lip 1:1 N2;

- MP100+G+YE+Ac Chla:Lip 1:1 N2.

In the last step wheat straw was used (WS). The final aim was focused on growth conditions

applicable also with lignocellulosic biomass hydrolysates or other low cost feedstocks. The

samples were previously processed using acid (A) and acid plus enzymatic hydrolysis (AE).

Several different dilution factors (0, 2, 4) were used to verify inhition effects due to phenols

and furans. Finally, the cultures were inoculated by Lipomyces starkeyi DBVPG 6193 (Lip)

and microalga strain, Chlamydomonas sp. (Chla), separately and simultaneously (Chla:Lip

1:1) with the same inoculum concentrations of 5% v/v. All the tests were carried out, at least,

in duplicate. The samples were labeled as:

- WSAH Chla:Lip 1:1 dil 0, 2, 4 (Wheat Straw subjected to Acid Hydrolysis, using

Chlamydomonas:Lipomyces starkeyi ratio of 1:1 and dilution factors of 0, 2, 4);

- WSAELH Chla:Lip 1:1 dil 0, 2, 4 (Wheat Straw subjected to Acid Hydrolysis

plus Enzymatic hydrolysis on the Liquid fraction, using

Chlamydomonas:Lipomyces starkeyi ratio of 1:1 and dilution factors of 0, 2, 4);

- WSAESH Chla:Lip 1:1 dil 0, 2, 4 (Wheat Straw subjected to Acid Hydrolysis

plus Enzymatic hydrolysis on the Solid fraction, using Chlamydomonas:Lipomyces

starkeyi ratio of 1:1 and dilution factors of 0, 2, 4).

2.5 Combined yeast and microalgal consortium in pilot scale raceway pond for

urban wastewater treatment and potential SCO production

In order to monitore the growth conditions of oleaginous yeast (Lipomyces starkeyi DBVPG

6193) and microalgal consortium in mixed cultures using urban wastewater as different low

cost feedstock, the experimental activity was carried out in a raceway pond.

36

The goal of these tests was focused also on the decrease of capital costs strictly related to the

operating conditions (no sterile) and to the reactor design (raceway pond).

The tests were carried out in a 200 L raceway pond operating outdoor with one single-loop

open channel and semi-circular end-walls. The pilot plant was installed on the roof of the

Department of Bioscience and Territory, University of Molise, Pesche (IS), Italy [Figure 2.1].

It operated during the month of July with an average natural light intensity of approximately

600 µmol m-2

s-1

and natural light/dark cycles. A four-blade paddle wheel driven by a motor

engine working at 6 rpm supported the mixing of the culture media. The culture media

consisted in untreated urban wastewater, half diluted with tap water for a total volume of 150

L.

The microalgal consortium, used as inoculum, was obtained from the urban wastewater

treatment plant of Isernia (Italy) and maintained in laboratory controlled condition using

BBM medium in 1 L flask in agitation (150 rpm) to maintain the biomass in suspended

condition, under a continuous light of 1500 Lux (Cool White Fluorescent Lamps) at 25 °C.

Optical microscope analysis showed that inoculum was composed by cyanobacteria, diatoms

and microalgae (mostly Scenedesmus sp. and Chlorella sp.).

Lipomyces starkeyi DBVPG 6193 was added as oleaginous yeast following the culture

conditions as already described [Chapter 2, § 2.3]. Several physiological studies relating to

growth and lipid production by Lipomyces starkeyi was reported in literature but urban

wastewater was tested as growth medium for this microorganism (Tapia et al., 2012). The

microalgal consortium and Lipomyces starkeyi inoculum concentration was 3% and 1.5% v/v,

respectively. The cultivation was carried out for 14 days in batch mode. The sample was

labeled as:

- UWRP Lip:m.c (Urban Wastewater in Raceway Pond, using Lipomyces starkeyi and

microalgal consortium ad inoculum).

To identify bacteria, mould and fungus colonies, culture media were used to inoculate Petri

dishes with following media:

Brain Heart Infusion agar (BHI);

Mannitol Salt agar (MSA);

Mac Conkey agar (MAC);

Sabouraud broth (SAB).

37

Fig. 2.1: Pilot scale raceway pond installed on the roof of the Department of Bioscience and Territory, University of Molise,

Pesche (IS), Italy

2.6 Analytical methods

Measurements of pH were made by a inoLab® Multi 740 Multimeters pH-meter (WTW). The

biomass concentration was monitored with a Shimadzu UV6100 and Thermo Scientific

Helios Epsilon spectrophotometer (Japan) and by measuring turbidity of liquid samples at 600

nm. Dry cell weight (DCW) or microbial biomass (g/L) were determined by filtering 2-3 mL

of culture over pre-weight PES filters ( .45 μm; Sartorius Biolab, Germany). The retained

biomass on filters was washed, dried at 105 °C for 24 hours and then stored in a desiccator

before being weighed. The individual cell counts of yeast and microalga was determined with

a haemocytometer. When the microorganisms were cultured in SRF, turbidometric

measurements could not be carried out due to the darkness of the medium. Consequently, the

total count of microorganisms was carried out by sequential diluition in Petri dish containing

YPD agar medium (peptone 10 g/L, yeast extract 10 g/L, glucose 20 g/L, agar 15 g/L)

(Sigma-Aldrich). The cell proliferation was measured as CFU/mL (Colony Forming

Unit/mL).

After centrifugation and filtration with .2 μm cut-off filters, the liquid sample was analysed

for residual substrate content (glucose) and soluble fermentation products (VFA, alcohols).

Glucose was measured using an enzymatic kit (Sigma Aldrich). Reducing sugars and pentose

sugars were measured by Nelson-Somogyi assay (Sadarivam and Manickam, 1996) and

Fluoroglucinol (Douglas, 1981) colorimetric assay, respectively. Glucose was also analyzed

by HPLC (LC2010, Shimadzu, Japan), equipped with a refractive index detector (RID-20A,

Shimadzu, Japan).

The total concentration of phenolic compounds was determined using Folin-Ciocalteu

(Singleton et al., 1999), using cathecol as standard. A simple method based on UV spectra

was followed for the estimation of total furans (furfural and hydroxymethylfurfural) in the

38

hydrolysates (Martinez et al., 2000). UV spectra were recorded on a spectrophotometer using

1 cm cells.

COD was analysed by a colorimetric method using Hach vials. The TOC measurements were

carried out with a TOC-VCSH/CSN (Shimadzu, Japan), upon suitable dilution of a culture

medium sample. The TOC values were obtained subtracting the IC (inorganic carbon) value

from the TC (total carbon) value.

For volatile organic acids, alcohols and carbohydrates were also used the following methods:

Volatile organic acids and ethanol were determined by GC analysis, using a Shimadzu

GC-17A equipped with a FID detector and a capillary column with a PEG stationary

phase (BP2 , 3 m by .32 mm i.d., .25 μm film thickness, from SGE). Samples of 1

μL were injected with a split-ratio of 1:10. Helium was fed as carrier gas with a flow

rate of 6.5 mL/min. Injector and detector temperatures were set to 320 °C and 250 °C,

respectively. Initial column temperature was set to 30 °C, kept constant for 3 min,

followed by a ramp of 10 °C/min till 140 °C, then kept constant for 1 min.

VFA (Volatile Fatty Acids) were also analyzed by HPLC (LC2010, Shimadzu, Japan),

equipped with a refractive index detector (RID-20A, Shimadzu, Japan).

Concentrations of organic acids, alcohols and carbohydrates were measured by HPLC

(Dionex Ultimate 3000) with a refractive index detector (Waters R410). Samples were

first centrifuged at 12,000 g for 15 min and then supernatants were filtered with 0.2

µm syringe filters. HPLC analysis were performed at a flow rate of 0.7mL/min on an

Aminex HPX-87H, 300 x 7.8 mm (Bio-Rad) column at a temperature of 35 °C. H2SO4

at 4 mM was used as the mobile phase.

NH4+, NO3

ˉ and PO4

3 ˉ content were measured in the soluble phase by ion chromatograph

(ICS 3000, Dionex) equipped with pre-columns and separation columns CG 16 and CS16 (3

mm ø) for cations and AG15 and AS15 (2 mm ø) for anions, respectively. The eluents used

for this analysis were methanesulphonic acid (MSA) 25 mM at 0.35 mL/min for cations and

KOH 10 mM at 0.3 mL/min for anions at initial time and MSA 40 mM at 24 min and KOH

74 mM at 28 min. The column temperature was set at 35 °C. To obtain the soluble fractions,

the samples were centrifuged at 14,000 rpm for 10 min (Eppendorf 5424).

Lipid estimation was performed with sulfo-phospho-vanillin assay (Mishra et al., 2014).

Lipids were extracted by a method adapted from Bligh and Dyer (Bligh and Dyer, 1959). The

samples were stirred in a CHCl3/CH3OH mixture (2:1 w/v) over 24 hours, and the oleaginous

biomass was filtered off and washed with additional CHCl3. This procedure was repeated

three times. The solvent was then removed by evaporation under N2 stream.

39

The total lipid concentration was estimated by gravimetric method. To calculate the lipid

concentration of the cells, the cells were dried to a constant weight with an oven at 80 °C. The

lipids extracted were subjected to transesterification reaction in a stirred container at 60 °C for

10 min, using NaOH (1% w/v) or HCl (1% w/v) as catalysts and using methanol as reagent.

The samples were dried by N2 stream and subsequently 1 mL of heptane was added for the

analysis. The fatty acid compositions of the FAME were analyzed using gas chromatography (GC). The GC (GC-MS 2010, Shimadzu, Japan) was equipped with a flame ionization

detector and an Omegawax 250 (Supelco) column (30 m x 0.25 mm I.D., 0.25 µm). Helium

was used as carrier gas (flow rate: 30 mL/min). The samples were initially dissolved in 1 mL

of heptane and 1 µL of this solution was loaded onto the column. The temperature of the

column was kept at 50 °C for 2 min, then heated to 220 °C at a rate of 4 °C/min, and finally

kept constant for 2 minutes. Methyl decanoate was used as internal standard. The peaks of

each methyl ester was identified by comparing the retention time with the peak of the pure

standard compound. Gas measurement for carbon dioxide (CO2), nitrogen (N2) and oxygen (O2) was performed

using Perkin Elmer® Clarus 480 Gas Chromatograph, Waltham, USA. 200 µL of each sample

was injected in a capillary R-Q column separating CO2 from other gases. The remaining gases

were separated by another capillary Rt-Molsieve 5Å column. The temperature of the injector

was 250°C and the thermic conductivity detector was at 150 °C and the vector gas was Argon

(350 kPa, 34 mL/min). Calibration was performed on this machine with a standard gas with

25% CO2, 0.1% H2S, 0.5% O2, 10% N2 and 64.4% CH4 in composition. Biogas measurement

was performed for measuring gas composition in the reactors once a week after steady state.

A fluorometer (AquaFluorTM; Handheld Fluorometer/Turbidimeter; Turner Designs) was

used to measure the content of in vivo chlorophyll a in the untreated samples. Therefore the

excitation light of the fluorometer passes through the medium and causes the Chlorophyll a

inside the cells to fluoresce. This signal allows estimating the concentration of Chlorophyll a

in the sample.

40

3. RESULTS AND DISCUSSION